1. Field of the Invention
The present invention relates to a heterojunction bipolar transistor, more specifically, it relates to a bipolar transistor with a heterojunction, formed from, for example, gallium arsenide (GaAs) and aluminum gallium arsenide (AlGaAs), in which the emitter efficiency is improved. The present invention also relates to a process for fabricating such a heterojunction bipolar transistor.
2. Description of the Related Art
Recently, heterojunction bipolar transistors have been subjected to intense research and development in order to discover a method for obtaining a higher speed of operation.
For example, a frequency demultiplier comprising GaAs field effect transistors (FET's) or high electron mobility transistors (HEMT's) are operative at cut-off frequencies of up to 4 to 5 GHz. On the other hand, it has been reported that a frequency demultiplier comprising heterojunction bipolar transistors could operate at cut-off frequencies higher than 8 GHz at room temperature. Moreover, a cut-off frequency F.sub.T of more than 40 GHz has been attained by an AlGaAs/GaAs heterojunction bipolar transistor. On the other hand, the cut-off frequency of a silicon transistor is 20 GHz at highest.
An example of an AlGaAs/GaAs heterojunction bipolar transistor in the prior art has the following structure. On a semi-insulating GaAs substrate, an n.sup.+ -type GaAs collector contact layer, an n-type GaAs collector layer, a p.sup.+ -type GaAs base layer, an n-type Al.sub.x Ga.sub.1-x As grading layer (wherein x=0 to 0.3), an n-type Al.sub.0.3 Ga.sub.0.7 As emitter layer, and an n.sup.+ -type GaAs emitter contact layer are consecutively grown. Then a first etching is conducted to form a mesa comprising the emitter contact layer, the emitter layer, and the grading layer, and to expose a portion of the base layer. Beryllium (Be) or magnesium (Mg) ions are implanted into the exposed portion of the base layer, which is then activated by annealing to form a p.sup.+ -type base contact region therein.
A second etching is conducted to form another mesa comprising the above mesa and the base layer, and to expose a portion of the collector contact layer. On the base contact region, a base electrode is formed. An emitter electrode is formed on top of the mesa (i.e., the emitter contact layer), and a collector electrode is formed on the exposed portion of the collector contact layer. Thus, the main parts of a heterojunction bipolar transistor are constructed.
In this heterojunction bipolar transistor, the forbidden band gap of the AlGaAs emitter layer is larger than that of the GaAs base layer. Therefore, this type of heterojunction bipolar transistor is called "a wide gap emitter bipolar transistor". Essentially, a wide gap emitter bipolar transistor may have a higher emitter efficiency than a homojunction bipolar transistor, since holes (minority carriers for the emitter) flowing from the base to the emitter are blocked by the higher barrier in the valence band.
In practice, however, the emitter efficiency is relatively reduced by a recombination current generated in the emitter-base junction depletion layer. The current gain h.sub.FE of the heterojunction bipolar transistor is affected by the recombination current.
Generally, the recombination current depends on the quality of the crystals of the emitter and base layers, and the x value and the thickness of the Al.sub.x Ga.sub.1-x As grading layer, etc.
Specifically, in the heterojunction bipolar transistor having the structure described above, the emitter-base junction has a region exposed outside the crystal layers, whereat surface states, contamination, damage and the like are easily formed, causing the density of the recombination center to become high, and thus making recombination current generation high.
In the above heterojunction bipolar transistor, it is necessary to etch a portion of the emitter contact layer and the emitter layer to expose a portion of the base layer, to effectively reduce the base resistance. If the above etching is not effected, a base electrode must be formed on the emitter contact layer or on the emitter layer, and the base electrode and the base layer must be electrically connected through a region of the AlGaAs layer by doping a p-type impurity therein. If the base electrode is formed on the AlGaAs layer, the contact resistivity therebetween is relatively high. This is because the barrier height of the metal-AlGaAs junction is higher than that of the metal-AlGaAs junction due to a wider bandgap of AlGaAs. Another reason is that an oxide is easily formed at the surface of the AlGaAs layer, which acts as an insulator between the metal and AlGaAs. If the base electrode is formed on the emitter contact layer by doping a p-type impurity in a region of the emitter region to be converted there to a p-type conductivity in order to prevent the above mentioned effect, electrons and holes will flow through the junction between the n-type and p-type top GaAs regions, which reduces the emitter efficiency. As a result, exposure of the periphery of the interface between the base layer and the emitter layer, or specifically the grading layer, is essential for the structure of the heterojunction bipolar transistor described above, to reduce the base resistance.